Deaths after intravenous regional anaesthesia.

acne. Neither the severity of the acne nor the rate of excretion of sebum was measured, but the controls were well matched. Nevertheless, 10 patients had irregular menstrual periods and 12 had hirsuties, which may suggest that many of these 38 patients had clinical evidence of abnormal androgen function. The clear message is that we do not fully understand why patients with acne do have an increased sebum excretion rate; more basic research is needed. We do know, however, that a substantial fall in the rate of excretion of sebum is always associated with improvement in the acne. How much the secretion of sebum must be suppressed to produce clinical improvement is uncertain, but a reduction of 3500 or more is probably needed to produce a satisfactory clinical response. An oral contraceptive pill containing either 20 or 30 ,ug of oestrogen usually has no real effect on the woman's acne. A contraceptive pill with 50 Mtg of oestrogen will reduce the rate of excretion of sebum by 400o, and this is usually associated with clinical improvement.12 The combination of ethinyloestradiol 50 yg with prednisone 5 mg (the latter taken at night to obtain maximum adrenal suppression) will reduce the rate by up to half, with clinical improvement in almost all cases.13 The most potent combination of all is probably ethinyloestradiol 50 ,tg and the antiandrogen cyproterone acetate (2 or 100 mg), which will reduce production of sebum by 50-75%o. The treatment is given in the reverse sequential manner14that is, the oestrogen is taken from the fifth to the 25th day and the antiandrogen from the fifth to the 14th day. Considerable improvement in the acne may be expected with this regimen, but (as with all the hormonal regimens) not until about the sixth week of treatment. The production of sebum may also be reduced by up to 90%o without hormonal treatment: oral 13-cis-retinoic acid, which affects sebaceous gland differentiation, is most effective in treating patients with antibioticresistant acne.15 No successful topical antiseborrhoeic treatment is available; several formulations have been tried without effect. Their failure may be due either to poor percutaneous absorption or to a metabolite of the antiandrogen being required to mediate the desired effect. Alternatively, the plasma concentration of androgen reaching the sebaceous gland may easily overwhelm the local antiandrogen effect. In practical terms, which patients should be considered for hormonal treatment? The answer is only those who fail to respond to adequate conventional treatment, since it has fewer side effects than treatment with either hormones or retinoids. Conventional treatment means a minimum of 1 g a day of oral tetracycline or erythromycin together with topical preparations such as benzoyl peroxide given for a minimum of four to six months. Only a few patients with acne-about 2-5% -will need to be considered for treatment with hormones or retinoids. 13-Cisretinoic acid, now available on a named-patient basis, may be used in both sexes, whereas hormonal treatment is usually indicated only for women. Should a patient with acne not respond well to conventional treatment after three to four months and already be taking the contraceptive pill, then (if there are no contraindications) its ethinyloestradiol content should be increased to 50 ,tg. If after a further three months there is no response, or if the patient is not already taking the contraceptive pill, a hormone combination should be prescribed: either 50 ,.g ethinyloestradiol and 5 mg oral prednisone or the ethinyloestradiol-cyproterone acetate combination, which is now available in a twin pack. Finally, we need to remember that comparative studies of these three regimens-conventional treatment, hormones, or retinoids-have not been reported; until these have been performed absolutely reliable therapeutic guidelines cannot be given. W J CUNLIFFE